The Section conducts studies to elucidate the molecular and biochemical mechanisms of heritable disorders of connective tissue, specifically osteogenesis imperfecta (OI) and Ehlers-Danlos Syndrome (EDS), and to apply this information to the treatment of those disorders. We have previously developed a knock-in murine model for the non-lethal form of OI, which we have named the Brittle mouse (Brtl). Brtl has a classic glycine substitution (G349C) in one of its collagen alpha 1(I) chains and reproduces the molecular, biochemical and histological features of OI. We have been engaged in collaborative studies of the histomorphometry and biomechanics of Brtl long bones. At the midshaft, Brtl bone has a decreased cross-sectional area at 1,2, and 6 months of age and the calculated moment of inertia is reduced, predicting that Brtl bone should be weaker based on its geometric properties. In 4-point bending tests to failure, Brtl bone breaks with a lower force than controls at 1 and 2 months but has equal strength at 6 months. This combination of weak geometric properties and equal mechanical properties predicts that the composition of Brtl bone itself has changed after puberty in the mouse. This makes the Brtl mouse an excellent model for the well-described clinical feature of OI, decreased bone fragility after puberty. We are also using the Brtl mouse to conduct treatment trials of bisphosphonate. The treatment trials in the mouse are paralleled by treatment trials in children with types III and IV OI. In the pediatric trial, we are doing a four-arm controlled trial in which children are randomly assigned to pamidronate, growth hormone, both drugs or no drug. The effects of the treatment on vertebral bodies and on long bones are being compared. In the murine trial of the drug, we are using the bisposphonate alendronate. Controls or Brtl mice have already shown increased bone density of spine and long bone in response to treatment. Biomechanical testing will show how this increased density affects bone strength and brittleness. The Brtl mouse is also an excellent model in which to approach the gene therapy of OI, because a ribozyme (RZ) cleavage site was engineered into the Brtl mutant allele. Cleavage of mutant transcript by RZ can suppress expression the mutant protein and biochemically mimic a null allele, which is clinically mild in humans. In the past year, we have published our work with RZ in cultured OI fibroblasts, showing allele-specific suppression of mutant collagen transcript to about 50% of baseline levels in control cells. We generated a RZ mouse for transferring the therapeutic molecule by matings with Brtl and are in the process of testing the effect of the RZ on normal bones of the RZ transgenic and of comparing the anti-sense effects of various RZ constructs. In our studies of human collagen mutations causing OI, we have been focusing on a set of mutations in types I and V collagen that affect fibril formation. In one these mutations, there is a deletion of the exon in a1 (I) that contains the telepeptide binding site. Absence of this exon in even a small fraction of monomers results in a dramatic delay in fibril formation for the proband?s total collagen mix. Since this delay, primarily the longitudinal portion of fibril assembly kinetics, the collagen fibrils formed in vitro are 4-5 times the length of control molecules.